Anyone happen to know of an archive of coreimage kernel code for the standard comping modes (darken, lighten, multiply, screen etc.)?
I'm looking to take 2 image inputs and combine them in different ways.
Also, does anyone know if it is possible to use cikernel coding, to group, for example, all the pixels of one line of an image together, then do stuff to that one line (ie offset it's position). All the cikernel code I've seen so far seems to apply the same operation to all the pixels of the input image(s). I'd quite like to emulate the old 'scan processing' analogue video effects by drawing each frame a line at a time, then doing stuff to each line as it's scanned across. Or, group the image into blocks, and randomly shuffle them around.
Maybe not the right place to ask, but just wondering if anyone has tried something similar using Quartz Composer.
alx
http://www.toneburst.net